Creating floor plans for large areas via manual surveying is labor-intensive and error-prone. In this paper, we present a system, Walk&Sketch, that creates floor plans of an indoor environment by a person walking through the environment at a normal strolling pace and taking videos using a consumer RGB-D camera. The method computes floor maps represented by polylines from a 3D point cloud based on precise frame-to-frame alignment. It aligns a reference frame with the floor and computes the frame-to-frame offsets from the continuous RGB-D input. Line segments at a certain height are extracted from the 3D point cloud, and are merged to form a polyline map, which can be further modified and annotated by users. The explored area is visualized as a sequence of polygons, providing users with the information on coverage. Experiments have done in various areas of an office building and have shown encouraging results.

When a robotic team is deployed to provide sensing and communication support for human activities, it is crucial that the robotic team be able to not only react to the current perceived location of the human of interest, but also predict his/her future locations. In this work we present a novel approach to predicting human motion in indoor environments. Our approach consists of an offline pre-processing phase, in which we analyze a map of the indoor environment and identify distinct areas of interest, such as rooms, and a run-time phase, in which we maintain hypotheses of destinations and probabilistically update these hypotheses based on observations of the human's motion. Our approach creates a probability distribution over future locations for the human user at each time step in the future. Such predictions can be used for the robotic team to effectively track and relay information from the human user to the base station.

Mapping is important for mobile robots in unknown environments. However, existing work largely relies on long-range sensors such as expensive laser range finders and cameras. Little has been done for robots with only short-range sensors such as bumpers or wall sensors. In this paper, we fill in the gap and propose a method for real-time indoor mapping for blind robots, i.e. robots that have only short-range sensors. Several key problems for mapping are addressed in this paper: (1) generating map from a single robot's odometry and short-range sensor data, (2) combining and collating maps from multiple robots to generate a coherent bigger map, and (3) detecting loops in the map and performing loop closure. The method has been implemented on iRobot Create and tested in hallways of office buildings. Experimental results have shown that the system is robust to noisy odometry and obstacles along the walls.

Mapping and localization for indoor robotic navigation is a well-studied field. However, existing work largely relies on long range perceptive sensors in addition to the robot's odometry, and little has been done with short range sensors. In this paper, we propose a method for real-time indoor mapping using only short range sensors such as bumpers and/or wall sensors that enable wall-following. The method uses odometry data from the robot's wall-following trajectory, together with readings from bumpers and wall sensors. The method first performs trace segmentation by fitting line segments to the noisy trajectory. Given the assumption of approximately rectilinear structure in the floor plans, typical for most indoor environments, a probabilistic rectification process is then applied to the segmented traces to obtain the orthogonal wall outlines. Both segmentation and rectification are performed on-line onboard the robot during its navigation through the environment. The resulting map is a set of line segments that represents the wall outline. The method has been tested in office buildings. Experimental results have shown that the method is robust to noisy odometry and non-rectilinear obstacles along the walls.

Jibbigo is a speech-to-speech translation application for iPhone, iPod touch, and iPad devices. Jibbigo allows the user to simply speak a sentence, and it speaks the sentence aloud in the other language, much like a personal human interpreter would. The speech-to-speech translation is bidirectional for a two way dialog between participants.

Abstract — In this paper, we propose and study a novel dis-tributed vehicle traffic information system using a wireless sensor network. Besides the system model, the contributions of the paper include a road network aware information publication protocol, a distributed query processing protocol with transient memory, and the adaptive interaction model between the two, all in the context of a vehicle traffic information system. Both theoretical and simulation results are presented. I.

Many applications of sensor networks require the base station to collect all the data generated by sensor nodes. As a consequence many-to-one communication pattern, referred to as convergecast, is prevalent in sensor networks. In this paper, we address the challenge of fast and reliable convergecast on top of the collision-prone CSMA MAC layer. More specifically, we extend previous work by considering the following two situations: (1) the length of the packets generated by nodes is much smaller than the maximum length of a data frame that can be transmitted in one time slot and (2) not every node in the network has data to transmit and for those that have, may have lots of data that require more than one packet. The first situation leads to the possibility of improvement by data piggybacking/aggregation; the second scenario arises in networks where nodes locally store the data and serves query request on-demand. We present distributed minimal time scheduling algorithms for both the cases. Simulation results have shown significant performance improvements of our new approaches over existing solutions.

In this paper, we investigate efficient strategies for supporting on-demand information dissemination and gathering in large-scale wireless sensor networks. In particular, we propose a "comb-needle" discovery support model resembling an ancient method: use a comb to help find a needle in sand or a haystack. The model combines push and pull for information dissemination and gathering. The push component features data duplication in a linear neighborhood of each node. The pull component features a dynamic formation of an on-demand routing structure resembling a comb. The comb-needle model enables us to investigate the cost of a spectrum of push and pull combinations for supporting query and discovery in large-scale sensor networks. Our result shows that the optimal routing structure depends on the frequency of query occurrence and the spatial-temporal frequency of related events in the network. The benefit of balancing push and pull for information discovery is demonstrated

This article proposes the minimum power configuration (MPC) approach to power management in wireless sensor networks. In contrast to earlier research that treats different radio states (i.e., transmission/reception/idle) in isolation, MPC integrates them in a joint optimization problem that depends on both the set of active nodes and the transmission power. We propose four approximation algorithms with provable performance bounds and two practical routing protocols. Simulations based on realistic radio models show that the MPC approach can conserve more energy than existing minimum power routing and topology control protocols. Furthermore, it can flexibly adapt to network workload and radio platforms.

There has been a lot of work on localization for sensor networks. Most localization schemes assume an already deployed network. Very little research has been done for active sensor deployment with mobile robots. In this paper, we present a distributed sequential localization process which is tightly integrated with robotic sensor deployment. Being location-aware, sensors can be placed in advantageous locations to avoid big localization errors. In turn, the accurate localization result helps further sensor deployment. We compare our distributed sequential localization scheme with five existing localization algorithms, both through simulations and through repeated runs on a mote testbed. Results show that the distributed sequential localization algorithm produces comparable results to the state-of-the-art centralized localization algorithms. Experiments also show that pre-post processing plays important roles at getting good localization results for all localization algorithms

One of the most common communication patterns in sensor networks is routing data to a base station, while the base station can be either static or mobile. Even in static cases, a static spanning tree may not survive for a long time due to failures of sensor nodes. In this paper, we present an adaptive spanning tree routing mechanism, using real-time reinforcement learning strategies. We demonstrate via simulation that without additional control packets for tree maintenance, adaptive spanning trees can maintain the “best” connectivity to the base station, in spite of node failures or mobility of the base station. 1.

We consider applications of sensor networks wherein data packets generated by every node have to reach the base station. This results in a many-to-one communication paradigm referred to as convergecast. We are interested in determining a TDMA schedule that minimizes the total time required to complete the convergecast. We consider a simple version of the problem wherein every node generates exactly one packet. We propose a distributed convergecast scheduling algorithm that requires at most 3N timeslots, where N represents the number of nodes in the network. Through extensive simulations, we demonstrate that actual number of timeslots needed is around 1.5N. In addition to time efficiency, we prove that our convergecast scheduling algorithm requires the nodes to buffer no more than two packets at any instance. We propose a sleep schedule that conserves more than 50% of the energy. We present simulation results for a real application scenario to show that our convergecast scheduling algorithm performs significantly better than existing convergecast algorithms.

This paper described our handheld two-way speech translation system for English and Iraqi. The focus is on developing a field usable handheld device for speech-to-speech translation. The computation and memory limitations on the handheld impose critical constraints on the ASR, SMT, and TTS components. In this paper we discuss our approaches to optimize these components for the handheld device and present performance numbers from the evaluations that were an integral part of the project. Since one major aspect of the TransTac program is to build fieldable systems, we spent significant effort on developing an intuitive interface that minimizes the training time for users but also provides useful information such as back translations for translation quality feedback.

Location knowledge of nodes in a network is essential for many tasks such as routing, cooperative sensing, or service delivery in ad hoc, mobile, or sensor networks. This paper introduces a novel iterative method ILS for node localization starting with a relatively small number of anchor nodes in a large network. At each iteration, nodes are localized using a least-squares based algorithm. The computation is lightweight, fast, and any-time. To prevent error from propagating and accumulating during the iteration, the error control mechanism of the algorithm uses an error registry to select nodes that participate in the localization, based on their relative contribution to the localization accuracy. Simulation results have shown that the active selection strategy significantly mitigates the effect of error propagation. The algorithm has been tested on a network of Berkeley Mica2 motes with ultrasound TOA ranging devices. We have compared the algorithm with more global methods such as MDS-MAP and SDP-based algorithm both in simulation and on real hardware. The iterative localization achieves comparable location accuracy in both cases, compared to the more global methods, and has the advantage of being fully decentralized.

This paper proposes the minimum power configuration (MPC) approach to energy conservation in wireless sensor networks. In sharp contrast to earlier research that treats topology control, power-aware routing, and sleep management in isolation, MPC integrates them as a joint optimization problem in which the power configuration of a network consists of a set of active nodes and the transmission powers of the nodes. We show through analysis that the minimum power configuration of a network is inherently dependent on the data rates of sources. We propose several approximation algorithms with provable performance bounds compared to the optimal solution, and a practical Minimum Power Configuration Protocol (MPCP) that can dynamically (re)configure a network to minimize the energy consumption based on current data rates. Simulations based on realistic radio models of the Mica2 motes show that MPCP can conserve significantly more energy than existing minimum power routing and topology control protocols.

Translating out-of-vocabulary (OOV) terms is a great challenge for the Cross-lingual Information Retrieval and Data-driven Machine Translation systems. Several approaches have been proposed to mine translations for OOV terms from the web, especially from pages containing mixed languages. In this paper, we propose a novel approach to automatically translate OOV terms on the fly through cross-lingual query expansion. The proposed approach does not require any web crawling and has achieved an inclusion rate of 95% and overall translation accuracy of 90%, outperforming state-of-the-art OOV translation techniques.

In this paper we investigate efficient strategies for supporting on-demand information dissemination and gathering in large-scale vwireless sensor networks. In particular, we propose a "comb-needle" discovery support model resembling an ancient method: use a comb to help find a needle in sands or a haystack. The model combines push and pull for information dissemination and gathering. The push component features data duplication in a linear neighborhood of each node. The pull component features a dynamic formation of an on-demand routing structure resembling a comb. The comb-needle model enables us to investigate the cost of a spectrum of push and pull combinations for supporting discovery and query in large scale sensor networks. Our result shows that the optimal routing structure depends on the frequency of query occurrence and the spatial-temporal frequency of related events in the network. The benefit of balancing push and pull for discovery in large scale geometric networks are demonstrated. We also raise the issue of query coverage in unreliable networks and investigate how redundancy can improve the coverage via both theoretical analysis and simulation. Last, we study adaptive strategies for the case where the frequencies of query and events are unknown a priori and time-varying.

In this paper we describe the components of our statistical machine translation system. This system combines phrase-to-phrase translations extracted from a bilingual corpus using different alignment approaches. Special methods to extract and align named entities are used. We show how a manual lexicon can be incorporated into the statistical system in an optimized way. Experiments on Chinese-to-English and Arabic-to-English translation tasks are presented.

Large amounts of information are embedded in natural scenes. Signs are good examples of natural objects with high information content. In this paper, we discuss problems in automatic detection and translation of text from natural scenes. We describe the chal1enges of automatic text detection and propose methods to address these chal1enges. We extend example based machine translation technology for sign translation and present a prototype system for Chinese sign translation. This system is capable of capturing images, automatically detecting and recognizing text, and translating the text into English. The translation can be displayed on a palm size PDA, or synthesized as a voice output message over the earphones.

Large amounts of information is embedded in the natural scenes. Signs are good examples of objects in natural environments which have rich information content. In this paper, we present our efforts in the automatic sign translation. We describe the challenges in the automatic sign translation and introduce the architecture of our current system for automatic detection and translation of Chinese signs. Two data-driven machine translation methods: Example Based Machine Translation (EBMT) and Statistical Machine Translation (SMT) are compared for the task of translating Chinese signs into English. We report the experimental results of both methods that are trained from a small bilingual sign corpus combined with a bilingual glossary. The experiment results indicate that EBMT generates more correct translations while SMT is better at inferring unseen patterns. We are currently working on developing a multi-engine machine translation system that can incrementally learn from the data and combine the results from EBMT and SMT.